I've got 30-odd hours invested in a model, and I simply cannot comprehend MOI well enough to get it completed. I'm looking for assistance, either paid or unpaid - your choice.

What I'm trying to model is fairly straightforward, and MOI has all the necessary function, but I just can't make it work. Attached is a picture of an earlier version, as 3D printed in stainless/bronze.

Bottom line: want to make a quick $100 via PayPal paid in advance?
Work with me on this model so I can finally get my product out the door (many months late).

It's not a complex object: a truncated cone, curved to 45-degrees along its center axis, and capped with a hemispherical end.
Total height = 50mm. The base diameter is 20mm, the upper diameter is 27.5mm, and the dome size the same, with no "edge" between dome and cone (blended or fileted together where they meet).
The base should be solid for the first 15mm of height, and the rest of the interior hollow, with a consistent wall thickness of 3mm (and no inverted triangles on the inside that cause sliders to make it solid). It needs a concentric roughly 9mm stepped-diameter hole from the base into the inside, where I melt in a brass insert.

And I need it locked, so I can't continue to screw it up every time I look at it it... I'm so frustrated with this it's making me nuts, and I'll happily pay to get this part Done.

Anyone want to get involved? I'm certain it's no more than an hour or two for anyone who knows how to use MOI well.

Can you take a look at this result and comment?

A great start, but you hit the same issue I've been fighting: if the dome is attached when you bend it, the hemisphere becomes distorted and no longer hemispherical. I first tried making the dome with a round fillet on the cone and got that issue, so I detached the "dome" and tried manipulating the cone separately, but then started having fits re-attaching it and making it all a combined shape.

Also, the height shrank when you bent it - the bent pommel should be 50mm tall.

I also tried extruding the shape inward to get the wall thickness, thinking to use a cylindrical plug to make the bottom 15mm solid, but MOI exported the whole interior to .stl with what I take to be inverted triangles, and the result, after fixing with netfabb or Kisslicer, is no longer a hollow object. I've attached exported .stl of the best I've managed - note that the interior is apparently invalid, so netfabb fills it in solid, and KS draws infill lines across it. Very frustrating. (Please ignore that the "plug" is not flush with the end - I could not get them to align vertically.)

(Sure would be nicer if MOI did alignment like TinkerCAD - TK makes it so much easier with fewer clicks than MOI.)

I also included the .3dm that generated the invalid interior. It also has a couple of other obscure issues - the "ring" at the top, and the parallel lines along the outside curve - I cannot figure out why they are there, and they are visible in the printed .stl.

Ok, so lets look at this one.

You'll have to let go of the other files and errors. If you want to get into the misalignment of the generators and such, we can, after we get your part you want.

Also, all the "MoI cant move and align" and such, is just not correct. We can ask questions and learn about it after we get the part you want created. It most certainly CAN make precise moves and aligns.

Hello,

Am I off base in that you want something like this? It's not exact, just drew it up quickly. I don't really like the continuity at the top surface and the bunching at the bottom, but should be able to tweak that.

Wow! - this is more progress than I made in weeks - you obviously know the product well.

BTW, I wasn't griping, just mentioning the approaches I had tried. I should have been more clear - it's not that MOI can't do these things, it's just that I haven't been able to puzzle them out. I'll correct my comments about buggy Flow - it's not the software's fault, it's mine.

Draft-2 is excellent, but now I'm seeing dimensions issues - e.g, the base is now 19.73mm instead of 20, and the circle where the dome meets the cone appears to have departed from 27.5mm and round, although that could perhaps be due to measuring the tilted diameter horizontally? Also, the piece shrank vertically, and is now only 48mm tall - should be 50mm.

The hole needs to be added after bending, not before - it needs to be centered in the base, and straight and perpendicular to it, for the insert. As to the concentric hole size for the insert, here are the exact dimensions: http://www.mcmaster.com/#94180a373/=qm02my

I goofed on the wall thickness - should have said 2.5mm thick, not 3mm - if you could change that?

And that's about it - it's gorgeous otherwise.

OSTexo - I'm working with BurrMan on it - we're probably in the last revision, but yours is a nice start. I'm running a test print of Rev.2 from BurrMan now.

Sure nice to find experts on MOI around - I should have called out for assistance months ago.

Hello,

Understood, I'm making sure I understand the model so I can improve my skills, not looking to get in the middle of anything, e.g. trying different methods to get the end cap matching with the body of the part above G1.

I just ran it thru netfabb and Kisslicer for a test - it slices clean with no artifacts. I'm a bit chagrined that others have figured out how to do this, when I failed utterly...

Hi Rick,
Just so you know, the great point about doing it in the forum is others can help out.

I'm not so sure about what I'm making either. Would love to have others participate.

Like the "50 mm".... I've been having that be the "tip" ( the point at the top is 50 mm up the curve..

Here's another revision. This is using some sweeps and a blend to cap the end.

I have to run out a for a bit. We just need to get nailed down what you need. (The sizing off in the last part was just me and my generator curves. This can be set proper once we figure out what the part should look like, and where the dims you mention should be.

Also, yes. You should come in the forum long before you get frustrated. You'll find many users that can help.

If you can crop the to half sphere a little bit, you can end with a nice g2 continuity too ... but I'm not sure I completely get the problem either, language barrier :)

That's a really nice transition - enhances the shape quite a bit.

Hello,

PaQs model zebra lines. This part fits to the end of a fencing sword handle?

Here's a video of what I think PaQ is describing.



I'm still a little unclear on what some of the measurements are with regard to the geometry that was in the original files. Like the 27.5 part.

Anyway, I also showed a couple cuts to represent the "15 mm" cut in the bottom? Wasn't sure how that was supposed to go. If either of those are correct, let us know.

Once the shape is nailed down to what you want, we can discuss various ways for you to get it done.

[EDIT] in the video I just used a tangent blend. The file attached is what PaQ referred to, where I recut the sphere to be just a half sphere, then I could use a G2 blend so it was smooth.[EDIT]

Hello,

BurrMans zebra line part.

Yes - this is a new take on a Pommel - the part that screws onto the end of the tang of an epee, holds the French-type handle on, and serves as a counterweight for the blade. All previous pommels are basically tapered cylinders run on a lathe.
With a French grip, one is allowed to hold it anywhere along its length, but the tradeoff is less wrist-driven power than a "pistol" of orthopedic-type grip, vs. a couple of inches of extra reach and fencing with "absence of blade", i.e when the opponent tries to take control, the French-grip fencer simply evades blade contact, because manipulation with the fingers is faster than with the wrist. I've attached an image of the prototype I fence with.
Once the brass insert is melted into place, the fencer can then insert lead of tungsten shot, to get the exact balance desired for the epee - the other unique aspect - adjustable weight.

I think draft5-g2 may be just a bit too bulgy for the purpose, but I'll print it tonight and see how it feels. I think that the flare at the waistline is not going to work quite as well as the straight sides just merging smoothly into the dome. Needless to say this is very subjective, but I'm guessing the little finger won't fit quite as well on the bulge in 5-g2.

As to the 27.5mm - that is the diameter of the waistline where the cone meets the dome, and thus the diameter of the dome.
20mm is the diameter of the base.
The 15mm is the height of the solid portion at the bottom, which the (needs to be straight but dimensioned as at McMaster-Carr link) hole pierces.
Above the 15mm, the rest should be hollow, with 2.5mm wall thickness.

Getting real close. I'll do some test prints, and be back in the morning (we're shut down here in Portland, with an inch of freezing rain on top of 10" of dry snow).

Hey Rick,
So don't do a bunch of tests yet. Things like the "bulgy" aspect can easily be changed.

I think the main part first is to see if the "hollows and solids" are as they should be along with the basic dims. (For instance, I made the walls 3mm thick. You just said 2.5mm, so that would have to change still). Look at how the 15 mm solid is made and mention if that's correct. After those are all correct, we can look at a file where you get the shape you want. Even set you up to do some variations yourself. (Like note Paq's shape has not the same dramatic bulge) You'll see me set the sphere at an exact 45 angle. Is this non-negotiable? A slight rotation back on that would remove a good portion of that bulge, fairly easy.

Also, with regard to the 27.5 radius area (Where the "sphere meets the body"..... There will be a difference in that area being some sort of sweep or flow through those dimensions, and getting a curve of something "other than a perfect 27.5 arc", but that doesn't mean it's not conforming to those dims. The construction will be much more limited if you "must have an exact arc there".... Just a note.

Talk later.

This is my version :) Free of charge, of course. ;)

Just so you have something to reference, here is the object using the blend method with tangency. You can also note in this one I rotated the upper sphere a bit off of 45 degree's. This makes less "bulge", but maintains the basic gist of the part.

Anyway, talk later.


[EDIT] Foiled by Max!!!! [EDIT]

I'm currently test-printing Burr's v6, because a lot of the 3D printing design process is a matter of how the slicer and printer interpret the design - they are very interactive with the CAD. I've printed hundreds of these things while honing the process.

So far, so good - it's so close that I'm practically hyperventilating about releasing and shipping this product, the Melting Pommel, this week.

The exterior shape is perfect enough for this planet, the walls are 2.5mm, and it's printing really well on my TrinityLabs Aluminatus. http://trinitylabs.com/products/aluminatus-evo1 (Mine was the half-priced Beta version - considerable assembly and effort involved at the time)

I'm very excited to get this one done, as I'm supplying a couple of Rio 2014 Olympics candidates with pommels (this and the Palmel version linked above), and I believe that this one will become very popular, while stretching a couple of hidebound fencing rules to the limit and offering considerable advantage to fencers who use it.

BTW, this is pretty much a charitable activity on my part, as the market of Epeeists who fence Extended French style is very small, and very conservative. I'm producing these because I believe they can cause a significant change of balance in the sport, from advantage to those power-fencers who use a pistol grip toward those who use a French grip. So far I'm into it to the tune of about $5K, and chances of ever making that back (much less my time) are infinitesimal.

More detail on this version in an hour or two, but it's soooooo close.